Data storage method, device, system, electronic equipment and computer storage medium

CN115408143BActive Publication Date: 2026-07-07BEIJING KINGSOFT CLOUD NETWORK TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIJING KINGSOFT CLOUD NETWORK TECH CO LTD
Filing Date
2021-05-27
Publication Date
2026-07-07

AI Technical Summary

Technical Problem

In existing data storage systems, the corresponding threads become idle after a disk goes offline, leading to resource waste, and the thread allocation algorithm is complex.

Method used

A multi-threaded group architecture is adopted, with each storage disk corresponding to a thread group. A thread belongs to at least two thread groups at the same time. The target thread writes the contents of the main storage block to the backup storage block, ensuring that the thread can still handle other synchronous tasks after the disk goes offline, and simplifying the thread allocation algorithm.

Benefits of technology

It improves resource utilization, avoids thread idleness, achieves even distribution of synchronous tasks among threads, simplifies the allocation algorithm, and improves the system's resource utilization efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115408143B_ABST
    Figure CN115408143B_ABST
Patent Text Reader

Abstract

The present disclosure provides a data storage method, device, system, electronic equipment and computer storage medium. The data storage system comprises a processing device and a storage device, wherein the processing device runs a plurality of thread groups, each thread group comprising at least two threads, and the storage device comprises a plurality of storage disks, one storage disk corresponding to one thread group, and one thread belonging to at least two thread groups simultaneously. When synchronization of a target main storage block is required, a target thread corresponding to the target main storage block is obtained, the target thread being one thread in a thread group corresponding to the storage disk where the target main storage block is located, and the content of the target main storage block is written into a corresponding backup storage block through the target thread. In this way, each thread can accept a synchronization task corresponding to a main storage block in a different storage disk, and after any storage disk goes offline, the threads in the corresponding thread group can continue to complete synchronization tasks of other storage disks, thereby improving resource utilization.
Need to check novelty before this filing date? Find Prior Art